Increase in cognitive function is seen in many single-operated pediatric patients after epilepsy surgery

Open ArchivePublished:August 28, 2020DOI:https://doi.org/10.1016/j.seizure.2020.08.002

      Highlights

      • Epilepsy surgery allows for IQ gains.
      • Preoperative seizure frequency is a predictor of IQ change after surgery.
      • Single-operated patients have the best cognitive outcome after surgery.

      Abstract

      Purpose

      The recurrent seizures of pediatric drug-resistant epilepsy (DRE) are known to impair brain development and can lead to a loss in cognitive functioning. Surgery is increasingly being used to treat children with DRE. This study investigates the pre- and postoperative cognitive function in a pediatric epilepsy surgery cohort as well as predictive determinants of change in intelligence quotient (IQ) following surgery.

      Methods

      A consecutive series of 91 Danish children who underwent focal resective epilepsy surgery between January 1996 and December 2016 were included. All underwent preoperative cognitive evaluation and were reevaluated at 1-year and/or 2-year follow-up. Single-operated and multi-operated patients were examined separately.

      Results

      79 of 91 patients were single-operated. Single-operated patients received less anti-epileptic drugs (AED) and experienced a decrease in seizure frequency postoperatively, p < 0.001. IQ increased postoperatively (IQ change ± standard deviation: 3.3 ± 14.0), p < 0.05. High preoperative seizure frequency was a significant predictor for decreased IQ, p < 0.01. Multi-operated patients did not experience a reduction in AED treatment. Surgery and continued AED treatment did, however, result in significantly better seizure control, p < 0.01. IQ remained unchanged in multi-operated patients.

      Conclusion

      Epilepsy surgery allowed for IQ gains in single-operated patients. Preoperative seizure frequency was a significant predictor of IQ change following surgery. Interactions between other, not included, possible predictors remain to be examined. Single-operated patients had the best cognitive outcome. The inclusion of a non-surgical control group is needed to assess the extent of the beneficial effects of surgery on cognitive ability.

      Keywords

      1. Introduction

      Drug-resistant epilepsy (DRE) leads to a debilitating life with gross impact on quality of life [
      • Kwan P.
      • Brodie M.J.
      Early identification of refractory epilepsy.
      ]. Recurrent seizures are known to cause permanent and progressive changes in brain structure and function, leading to impaired brain development and a loss in cognitive functioning [
      • Bjørnaes H.
      • Stabell K.
      • Henriksen O.
      • Løyning Y.
      The effects of refractory epilepsy on intellectual functioning in children and adults. A longitudinal study.
      ,
      • Tromp S.C.
      • Weber J.W.
      • Aldenkamp A.P.
      • Arends J.
      • vander Linden I.
      • Diepman L.
      Relative influence of epileptic seizures and of epilepsy syndrome on cognitive function.
      ,
      • Matsuzaka T.
      • Baba H.
      • Matsuo A.
      • Tsuru A.
      • Moriuchi H.
      • Tanaka S.
      Developmental assessment-based surgical intervention for intractable epilepsies in infants and young children.
      ]. Cognitive development may slow or cease after the onset of seizures
      • Bjørnaes H.
      • Stabell K.
      • Henriksen O.
      • Løyning Y.
      The effects of refractory epilepsy on intellectual functioning in children and adults. A longitudinal study.
      ,
      • Tromp S.C.
      • Weber J.W.
      • Aldenkamp A.P.
      • Arends J.
      • vander Linden I.
      • Diepman L.
      Relative influence of epileptic seizures and of epilepsy syndrome on cognitive function.
      ,
      • Berg A.T.
      • Langfitt J.T.
      • Testa F.M.
      • Levy S.R.
      • DiMario F.
      • Westerveld M.
      Global cognitive function in children with epilepsy: a community-based study.
      ,
      • Gordon N.
      Cognitive functions and epileptic activity.
      ,
      • Bailet L.L.
      • Turk W.R.
      The impact of childhood epilepsy on neurocognitive and behavioral performance: a prospective longitudinal study.
      ,
      • Thompson P.J.
      • Duncan J.S.
      Cognitive decline in severe intractable epilepsy.
      ].
      DRE is defined by the International League Against Epilepsy as ‘failure of adequate trials of two tolerated and appropriately chosen and used AED schedules (whether as monotherapies or in combination) to achieve sustained seizure freedom’ [
      • Kwan P.
      • Arzimanoglou A.
      • Berg A.T.
      • Brodie M.J.
      • Allen Hauser W.
      • Mathern G.
      Definition of drug resistant epilepsy: consensus proposal by the ad hoc Task Force of the ILAE Commission on Therapeutic Strategies.
      ]. In such cases, other forms of timely intervention, which can achieve freedom from seizures and discontinuation of antiepileptic drugs (AED), is crucial in preserving, stabilizing and potentially improving the cognitive function of pediatric patients with epilepsy [
      • Bourgeois M.
      • Di Rocco F.
      • Roujeau T.
      • Boddaert N.
      • Lelouch-Tubiana A.
      • Varlet P.
      [Epilepsy and focal lesions in children. Su(1)rgical management].
      ,
      • Freitag H.
      • Tuxhorn I.
      Cognitive function in preschool children after epilepsy surgery: rationale for early intervention.
      ,
      • Stafstrom C.E.
      • Lynch M.
      • Sutula T.P.
      Consequences of epilepsy in the developing brain: implications for surgical management.
      ,
      • Holthausen H.
      • Pieper T.
      • Kudernatsch M.
      Towards early diagnosis and treatment to save children from catastrophic epilepsy -- focus on epilepsy surgery.
      ,
      • Ramantani G.
      • Kadish N.E.
      • Strobl K.
      • Brandt A.
      • Stathi A.
      • Mayer H.
      Seizure and cognitive outcomes of epilepsy surgery in infancy and early childhood.
      ,
      • Elger C.E.
      • Helmstaedter C.
      • Kurthen M.
      Chronic epilepsy and cognition.
      ,
      • Andersson-Roswall L.
      • Engman E.
      • Samuelsson H.
      • Sjöberg-Larsson C.
      • Malmgren K.
      Verbal memory decline and adverse effects on cognition in adult patients with pharmacoresistant partial epilepsy: a longitudinal controlled study of 36 patients.
      ,
      • Cormack F.
      • Cross J.H.
      • Isaacs E.
      • Harkness W.
      • Wright I.
      • Vargha-Khadem F.
      The development of intellectual abilities in pediatric temporal lobe epilepsy.
      ]. Surgical intervention is increasingly used to treat children with DRE. Pediatric epilepsy surgery aims to remove the epileptogenic area or limit the spread of seizure activity in the brain, thereby reducing seizure frequency and the need for AED treatment. Removal of the underlying epileptogenic pathology and reduction of AED treatment may then allow for recovery with improvements in cognitive functioning and neurodevelopment as a whole [
      • Gallagher A.
      • Jambaqué I.
      • Lassonde M.
      Cognitive outcome of surgery.
      ,
      • Van Schooneveld M.M.J.
      • Braun K.P.J.
      Cognitive outcome after epilepsy surgery in children.
      ,
      • Gleissner U.
      • Sassen R.
      • Schramm J.
      • Elger C.E.
      • Helmstaedter C.
      Greater functional recovery after temporal lobe epilepsy surgery in children.
      ].
      Many studies have documented a stabilizing effect of epilepsy surgery on seizure control [
      • Bourgeois M.
      • Di Rocco F.
      • Roujeau T.
      • Boddaert N.
      • Lelouch-Tubiana A.
      • Varlet P.
      [Epilepsy and focal lesions in children. Su(1)rgical management].
      ,
      • Freitag H.
      • Tuxhorn I.
      Cognitive function in preschool children after epilepsy surgery: rationale for early intervention.
      ,
      • Skirrow C.
      • Cross J.H.
      • Cormack F.
      • Harkness W.
      • Vargha-Khadem F.
      • Baldeweg T.
      Long-term intellectual outcome after temporal lobe surgery in childhood.
      ,
      • Wyllie E.
      • Comair Y.G.
      • Kotagal P.
      • Raja S.
      • Ruggieri P.
      Epilepsy surgery in infants.
      ,
      • Boshuisen K.
      • van Schooneveld M.M.J.
      • Uiterwaal C.S.P.M.
      • Cross J.H.
      • Harrison S.
      • Polster T.
      Intelligence quotient improves after antiepileptic drug withdrawal following pediatric epilepsy surgery.
      ,
      • Liu S.
      • An N.
      • Yang H.
      • Yang M.
      • Hou Z.
      • Liu L.
      Pediatric intractable epilepsy syndromes: reason for early surgical intervention.
      ,
      • Loddenkemper T.
      • Holland K.D.
      • Stanford L.D.
      • Kotagal P.
      • Bingaman W.
      • Wyllie E.
      Developmental outcome after epilepsy surgery in infancy.
      ,
      • Helmstaedter C.
      • Beeres K.
      • Elger C.E.
      • Kuczaty S.
      • Schramm J.
      • Hoppe C.
      Cognitive outcome of pediatric epilepsy surgery across ages and different types of surgeries: A monocentric 1-year follow-up study in 306 patients of school age.
      ]. The general cognitive level often remains stable following epilepsy surgery but with longer follow-up periods, intelligence quotients (IQ) have been shown to improve [
      • Matsuzaka T.
      • Baba H.
      • Matsuo A.
      • Tsuru A.
      • Moriuchi H.
      • Tanaka S.
      Developmental assessment-based surgical intervention for intractable epilepsies in infants and young children.
      ,
      • Freitag H.
      • Tuxhorn I.
      Cognitive function in preschool children after epilepsy surgery: rationale for early intervention.
      ,
      • Skirrow C.
      • Cross J.H.
      • Cormack F.
      • Harkness W.
      • Vargha-Khadem F.
      • Baldeweg T.
      Long-term intellectual outcome after temporal lobe surgery in childhood.
      ,
      • Loddenkemper T.
      • Holland K.D.
      • Stanford L.D.
      • Kotagal P.
      • Bingaman W.
      • Wyllie E.
      Developmental outcome after epilepsy surgery in infancy.
      ,
      • Westerveld M.
      • Sass K.J.
      • Chelune G.J.
      • Hermann B.P.
      • Barr W.B.
      • Loring D.W.
      Temporal lobectomy in children: cognitive outcome.
      ,
      • Roulet-Perez E.
      • Davidoff V.
      • Mayor-Dubois C.
      • Maeder-Ingvar M.
      • Seeck M.
      • Ruffieux C.
      Impact of severe epilepsy on development: recovery potential after successful early epilepsy surgery.
      ,
      • Wyllie E.
      Surgery for catastrophic localization-related epilepsy in infants.
      ]. However, the study cohorts are small and the majority include various pediatric patient groups with different epilepsy syndromes, various age ranges and varying cognitive abilities. Apart from limitations with small sample sizes, there is a general lack of randomized control trials with appropriate comparison groups. All these factors make it difficult to assess the effects of surgery on cognitive function.
      This is a retrospective follow-up study of a 20-year consecutive case series of Danish pediatric patients with DRE of different etiologies who underwent focal resection. The aims of this study were 1) to investigate the pre- and postoperative cognitive function in a pediatric epilepsy surgery cohort and 2) to identify predictive determinants of IQ change postoperatively.

      2. Materials and methods

      2.1 Participants

      This study comprised pediatric epilepsy patients from a consecutive series of 135 Danish, Greenlandic and Faroese children who underwent resective epilepsy surgery between January 1996 and December 2016 [
      • Pinborg L.H.
      • Jespersen B.
      • Beniczky S.
      • Fabricius M.
      • Rasonyi G.
      • Uldall P.
      Epilepsy surgery.
      ,
      • von Celsing Underbjerg E.
      • Hoei-Hansen C.E.
      • Madsen F.F.
      • Madsen C.G.
      • Høgenhaven H.
      • Uldall P.
      Danish experience with paediatric epilepsy surgery.
      ]. The cohort included patients with various medically intractable epilepsy syndromes and varying surgical interventions. Patients were considered eligible surgery candidates if they experienced incapacitating epileptic seizures with impaired awareness (≥ four seizures/month) and had been treated with a minimum of three AEDs, where a minimum of two AEDs had been tried simultaneously. Patients were referred from all of Denmark, Greenland and the Faroe Islands to either Copenhagen University Hospital Rigshospitalet or Epilepsy Hospital Filadelfia for preoperative evaluation and postoperative follow-up. Surgery was carried out at one of four locations – Copenhagen (Denmark), Cleveland (USA), Bielefeld (Germany) or Paris (France).
      Patients who underwent focal resection were either single- or multi-operated and were below 19 years of age at the time of first surgery. All included patients underwent both pre- and postoperative cognitive assessment. Patients were excluded if cognitive assessment could not be carried out satisfactorily, if there was a language barrier, if cognitive tests inappropriate for the mental age of the patient were used or if cognitive data was missing or lacking. Patients who underwent extensive resection, i.e. callosotomy and functional or anatomical hemispherectomy, were excluded from this analysis.

      2.2 Pre- and postoperative evaluations

      All candidates for epilepsy surgery were reviewed by a multi-disciplinary board including pediatric neurologists, neurophysiologists, neurosurgeons, neuroradiologists and neuropsychologists. Preoperative evaluation included clinical history, physical and neurological examination, continuous scalp video electroencephalogram (EEG) monitoring, 3 T high-resolution magnetic resonance imaging (MRI) and neuropsychological and cognitive assessments. When necessary, patients underwent additional intracranial EEG, functional imaging with interictal and/or ictal single photon emission computer tomography and positron emission tomography.
      Postoperative evaluations were carried out at 6 weeks and at 6, 12, 18, and 24 months after surgery. Additional follow-up consultations were carried out if needed. All standard postoperative evaluations included physical and neurological examination and registration of seizure frequency, postoperative complications and number of AEDs used. Repeated EEG and MRI were performed at 12 months. During the follow-up period, AED treatment was reduced on an individual basis.

      2.3 Cognitive assessments

      All patients underwent pre- and at least one postoperative cognitive assessment. Before 2005, postoperative cognitive tests were carried out at the one-year follow-up. Between 2005 and 2013 at both one- and two-year follow-up, and from 2013 at two-year follow-up.
      The choice of cognitive test used depended on the patient’s mental age and cognitive level. Depending on the preference of the assessing psychologist, either the Bayley Scale of Infant and Toddler Development or the Mullen Scale of Early Learning was used in patients up to age 70 months and in patients with severe mental retardation. A developmental quotient (DQ) equivalent to full-scale IQ was reported for these patients. Older patients and patients who were mentally capable were evaluated using the newest available version of the Wechsler Intelligence Tests. These included the Wechsler Preschool and Primary Scale of Intelligence (WPPSI) for children aged three to seven years, the Wechsler Intelligence Scale for Children (WISC) for children aged six to 18 years and the Wechsler Adult Intelligence Scale (WAIS) for children older than 16 or 18 depending on the test available at the time of testing. The main IQ outcome variables examined were preoperative IQ and the latest inventoried postoperative IQ. IQ change was analyzed in three groups, where IQ change (postoperative IQ minus preoperative IQ) was defined as decreased if there was a decline of ≥10 points, stable if the IQ change was -9 to +9 points and increased if the IQ change was ≥10 points.

      2.4 Data collection

      Retrospectively collected clinical and surgery-related data obtained from patient medical records were reviewed. For patients with multiple surgeries, MRI diagnosis, histological diagnosis, preoperative seizure frequency and preoperative AED treatment from the first surgery were registered. Regarding postoperative follow-up time, postoperative seizure frequency, postoperative AED treatment and postoperative IQ, data from the final surgery were registered. Preoperative IQ, age at surgery, time from admission into the surgical program to surgery and time from onset of seizures to surgery were registered for both the first and final surgery. Type of surgery, site of resection and postoperative complications were noted for all surgeries.
      Data on postoperative AED treatment were collected at the 24-month postoperative evaluation. Data on postoperative seizure frequency were collected from the latest follow-up consultation the patient attended, in many cases outside the surgery program’s final 24-month evaluation. Postoperative follow-up time was the time from surgery to the latest follow-up consultation the patient attended.

      2.5 Statistics

      Non-normally distributed variables were summarized as medians and interquartile range (IQR). Categorical variables were summarized as frequencies and proportions. The Kruskall-Wallis test was used to assess differences in non-normally distributed data. The χ [
      • Bjørnaes H.
      • Stabell K.
      • Henriksen O.
      • Løyning Y.
      The effects of refractory epilepsy on intellectual functioning in children and adults. A longitudinal study.
      ] test was used to test for differences in categorical data. Differences in IQ and number of AEDs pre- and postoperatively were analyzed using the Wilcoxon Signed Rank test. Due to small patient numbers, seizure frequency was not analyzed according to Engels classification. Differences in seizure frequency were analyzed using the McNemar test after seizure frequency was grouped into daily, less than daily and seizure-free categories. Univariable and multivariable linear regression was used to determine predictors for IQ change. We tested for multicollinearity to identify suitable covariates for multivariable linear regression. A p-value of <0.05 was considered statistically significant. All analyses were conducted using STATA Statistical Software: release 15. College Station, TX: StataCorp LP and IBM SPSS Statistics, Version 22.0; Armonk, NY: IBM Corp.

      2.6 Ethics

      The study was approved by The Danish Health and Medicines Authority (3−3013-1152/1) as well as The Danish Data Protection Agency (2013−41-2459), as is required by Danish law. The study complies with the 2nd Declaration of Helsinki.

      3. Results

      3.1 Study population

      A total of 91 of 135 consecutive patients who underwent resective epilepsy from January 1996 to December 2016 were included in the study, (Fig. 1). Of the included patients, 69 underwent surgery in Denmark while the remaining 22 patients received at least one of their surgeries overseas. Of the included patients, 79 were operated once, nine were operated twice and three were operated three times. The median age of epilepsy onset was 3.9 years and 2.4 years for single- and multi-operated patients, respectively. MRI and histological diagnoses were most frequently mesial temporal sclerosis and cortical dysplasia, (Table 1). The time from inclusion in the epilepsy surgery program to time of surgery was 1.1 years (range: 0.7–1.7 years) and 1.3 years (range: 0.6–2.2 years)/3.4 years (range: 2.6–5.3 years) for single- and multi-operated patients (primary surgery/final surgery) respectively. Postoperative complications were seen in 17 (21.5 %) single-operated patients and in four surgeries involving multi-operated patients (33.3 %). 68.0 % of the complications seen were temporary and resolved. No significant difference in IQ change between patients with and without complications was found for both single- and multi-operated patients.
      Table 1Patient characteristics and surgery related data for single- and multi-operated patients from a Danish pediatric epilepsy cohort study spanning from January 1996 to December 2016.
      Patient groupSingle-operated patients, n = 79Multi-operated patients, n = 12
      Male sex, n (%)45 (57.0)6 (50.0)
      Age at epilepsy onset, years (median (IQR))3.9 (1.5 to 7.3)2.4 (0.6 to 8.2)
      Age at surgery
      For multi-operated patients, data from primary and final surgery are noted.
      , years (median (IQR))
      12.1 (9.5 to 14.0)Primary surgery: 12.0 (3.0 to 15.0)

      Final surgery: 13.5 (6.8 to 17.0)
      Time from onset of epilepsy to surgery
      For multi-operated patients, data from primary and final surgery are noted.
      , years (median (IQR))
      6.1 (3.7 to 9.9)Primary surgery: 4.6 (2.5 to 9.6)

      Final surgery: 8.9 (6.1 to 11.4)
      Time from admission into surgery program to surgery
      For multi-operated patients, data from primary and final surgery are noted.
      , years (median (IQR))
      1.1 (0.7 to 1.7)Primary surgery: 1.3 (0.6 to 2.2)

      Final surgery: 3.4 (2.6 to 5.3)
      Laterality, right, n (%)32 (40.5)8 (66.7)
      MRI diagnosis
      For multi-operated patients, data from primary surgery is noted.
      , n (%)
      • -
        Mesial temporal sclerosis
      26 (32.9)0
      • -
        Tumor
      13 (16.5)2 (16.7)
      • -
        Arteriovenous malformation
      1 (1.3)0
      • -
        Cortical dysplasia
      16 (20.3)6 (50.0)
      • -
        DNET
      1 (1.3)2 (16.7)
      • -
        Infarction
      2 (2.5)0
      • -
        Sequelae from encephalitis, cysts, schizencephaly
      4 (5.1)0
      • -
        Sturge-Weber syndrome
      1 (1.3)0
      • -
        Hemimegaloencephaly
      01 (8.3)
      • -
        Normal
      6 (7.6)0
      • -
        Uncertain MRI findings#
      9 (11.4)1 (8.3)
      Type of operation
      For multi-operated patients, data for all surgeries are noted.
      , n (%)
      • -
        Amygdalohippocampectomy
      35 (44.3)0
      • -
        Focal resection
      44 (55.7)1. surgery: 12

      2. surgery: 12

      3. surgery: 3
      Site of resection
      For multi-operated patients, data for all surgeries are noted.
      , n (%)
      Temporal lobe53 (67.1)1. surgery: 8 (66.7)

      2. surgery: 4 (33.3)
      Frontal lobe11 (13.9)1. surgery: 3 (25.0)

      2. surgery: 5 (41.7)

      3. surgery: 1 (33.3)
      Parietal lobe3 (3.8)0
      Occipital lobe1 (1.3)2. surgery: 2 (16.7)
      Cerebellum1 (1.3)0
      Multilobar10 (12.7)1. surgery: 1 (8.3)

      2. surgery: 1 (8.3)

      3. surgery: 2 (66.7)
      Histological diagnosis
      For multi-operated patients, data from primary surgery is noted.
      , n (%)
      • -
        Mesial temporal sclerosis
      23 (29.1)0 (0.0)
      • -
        Tumor
      12 (15.2)2 (16.7)
      • -
        Arteriovenous malformation
      2 (2.5)0 (0.0)
      • -
        Hamartoma
      2 (2.5)0 (0.0)
      • -
        Cortical dysplasia
      25 (31.7)4 (33.3)
      • -
        DNET
      3 (3.8)2 (16.7)
      • -
        Other
      6 (7.6)3 (25.0)
      • -
        Normal
      5 (6.3)1 (8.3)
      • -
        Missing data
      1 (1.3)0 (0.0)
      Preoperative IQ
      For multi-operated patients, data from primary and final surgery are noted.
      , n (%)
      - ≤5025 (31.7)Primary surgery: 4 (33.3)

      Final surgery: 5 (41.7)
      - 51−6917 (21.5)Primary surgery: 3 (25.0)

      Final surgery: 2 (16.7)
      -70−8410 (12.7)Primary surgery: 1 (8.3)

      Final surgery: 1 (8.3)
      - ≥8527 (34.2)Primary surgery: 4 (33.3)

      Final surgery: 4 (33.3)
      Postoperative IQ
      For multi-operated patients, data from final surgery is noted.
      , n (%)
      - ≤5024 (30.4)4 (33.3)
      - 51−6910 (12.7)2 (16.7)
      - 70−8414 (17.7)2 (16.7)
      - ≥8531 (39.2)4 (33.3)
      Histology: dysplasia - three; normal findings - three.
      #Histology in single-operated patients: tumor tissue - two; dysplasia - three; atrophy - two; normal findings - two Histology in multi-operated patient: atrophy.
      Abbreviations: IQR: interquartile range; MRI: magnetic resonance imaging; DNET: dysembryoplastic neuroepithelial tumor; IQ: intelligence quotient. Tumor types included oligodendroglioma, astroglioma, astrocytoma, ganglioglioma. Other histological diagnoses include atrophy, fibrosis and gliosis. Of the cortical dysplasias, there were histologically four grade 1 and three grade 2 among single-operated patients and one grade 2 among the multi-operated. The remaining had unspecific cortical dysplasia.
      * For multi-operated patients, data from primary and final surgery are noted.
      ** For multi-operated patients, data from primary surgery is noted.
      *** For multi-operated patients, data for all surgeries are noted.
      **** For multi-operated patients, data from final surgery is noted.
      Excluded patients included 20 patients who underwent callosotomy or hemispherectomy and 24 focal resection patients who were excluded because of a language barrier or because of missing, lacking or flawed data, (Fig. 1). Of the 24 excluded focal resection patients, 18 were single-operated and six were multi-operated. Both excluded single-operated and multi-operated patients did not differ in gender, age at debut of epilepsy or age at primary or final surgery compared with included patients.

      3.2 Cognitive function of single-operated patients

      A total of 67 patients (84.8 %) underwent postoperative cognitive testing after one year, 45 patients (57.0 %) underwent postoperative cognitive testing after two years and 34 patients (43.0 %) underwent both one- and two-year postoperative cognitive testing. Postoperative IQ was calculated from the most recent postoperative IQ test. Furthermore, postoperative AED treatment was noted from the 24-month postoperative evaluation and postoperative seizure frequency was collected from the latest follow-up consultation, often later than 24 months after surgery.
      The number of patients with IQ over 84 was 34.2 %, (Table 1). A significant difference between pre- and postoperative IQ for all single-operated patients was found, p < 0.05. The mean ± SD change in IQ was 3.3 ± 14.0 points, (Table 2). The number of single-operated patients with a ≥10-point increase in IQ was 24 (30 %) and for 44 patients (56 %) IQ was stable (IQ change of -9 to +9).
      Table 2Pre- and postoperative IQ, AED treatment and freedom from seizures for single- and multi-operated patients.
      Single-operated patients, n = 79Multi-operated patients, n = 12
      PreoperativePostoperativep-valuePreoperativePostoperativep-value
      IQ, median (IQR)68.0 (47.0 to 90.0)76.0 (43.0 to 97.0)<0.05
      Mean ± SD change in IQ = 3.3 ± 14.0 points.
      60.5 (40.0 to 91.5)68.0 (40.0 to 88.8)0.46
      AED treatment, n (%)
      -0 AEDs1 (1.3)11 (14.7)<0.0010 (0.0)1 (8.3)0.56
      -1 AED11 (14.7)32 (42.7)3 (25.0)2 (16.7)
      -≥2 AEDs63 (84.0)32 (42.7)9 (75.0)9 (75.0)
      Seizure frequency, n (%)
      -Seizure-free0 (0.0)54 (68.4)<0.0010 (0.0)8 (66.7)<0.01
      -Not seizure-free
      Daily37 (46.8)5 (6.3)10 (83.3)2 (16.7)
      < Daily42 (53.2)20 (25.3)2 (16.7)2 (16.7)
      Postoperative collection time for IQ: latest available, either 12 or 24 months after final surgery; postoperative collection time for AED treatment: 24 months after final surgery; postoperative collection time for seizure frequency: from latest follow-up consultation, median (IQR) follow-up of 4.0 years (2.7–5.7) for single-operated patients and 3.3 years (2.6–4.8) for multi-operated patients.
      Abbreviations: IQ: intelligence quotient; IQR: interquartile range; AED: anti-epileptic drug.
      * Mean ± SD change in IQ = 3.3 ± 14.0 points.
      Patient characteristics for single-operated patients stratified by IQ change (postoperative IQ minus preoperative IQ) is given in Table 3. Preoperative seizure frequency and postoperative seizures were the only two factors that were significantly different between the groups, p = 0.001 and p < 0.05, respectively. The patient group with improved postoperative IQ was predominantly made up of patients suffering from less than daily seizures preoperatively, while the patient group with decreased postoperative IQ was predominantly made up of patients with daily preoperative seizures. Both the patient group with decreased postoperative IQ and the patient group with increased postoperatively IQ were predominantly seizure-free after surgery.
      Table 3Patient characteristics for single-operated patients, stratified according to IQ change (postoperative IQ-preoperative IQ) into three groups - decrease in IQ (≥10 points), stable IQ (-9 to +9 points) and increase in IQ (≥10 points).
      Single-operated patients
      Decrease in IQ

      (≥10 points)
      Stable IQ

      (-9 to +9 points)
      Increase in IQ

      (≥10 points)
      p-value
      Total, n (%)11 (14)44 (56)24 (30)
      Gender, n (%)
      • -
        Boys
      8 (18)25 (56)12 (27)0.48
      Laterality, n (%)
      • -
        Right
      5 (16)15 (47)12 (38)0.43
      Age at epilepsy onset, years (median (IQR))4.3 (0.1–5.5)3.9 (1.3–6.3)4.5 (2.1–8.4)0.28
      Age at surgery, years (median (IQR))9.7 (1.2–13.0)12.1 (10.0–14.0)13.1 (9.5–15.0)0.19
      Time from onset of epilepsy to surgery, years (median (IQR))4.1 (0.8–8.0)6.5 (4.5–10.2)5.4 (3.2–10.1)0.12
      Time from admission into surgery program to surgery, years (median (IQR))0.6 (0.5–1.6)1.1 (0.7–1.7)1.3 (0.7–1.7)0.21
      IQ change, median (IQR)−18 (-25 to -13)0 (-4 to 4)18 (13–23)
      Preoperative IQ, median (IQR)86 (65–95)60 (43–95)70 (56–84)0.22
      Postoperative IQ, median (IQR)56 (42–77)63 (40–96)90 (78–99)0.002
      Preoperative AED treatment, n (%)
      • -
        0
      0 (0)1 (2)0 (0)0.75
      • -
        1
      1 (9)8 (18)2 (8)
      • -
        ≥2
      10 (91)32 (73)21 (88)
      Postoperative AED treatment, n (%)
      • -
        0
      2 (18)5 (11)4 (17)0.56
      • -
        1
      5 (46)15 (34)12 (50)
      • -
        ≥2
      4 (36)21 (48)7 (29)
      Preoperative seizure frequency, n (%)
      • -
        < Daily
      1 (9)23 (52)18 (75)0.001
      • -
        Daily
      10 (91)21 (48)6 (25)
      Postoperative seizures, n (%)
      • -
        Seizure-free
      8 (73)25 (57)21 (88)0.03
      • -
        Not seizure-free
      3 (27)19 (43)3 (13)
      Postoperative collection time for IQ: latest available, either 12 or 24 months after final surgery; postoperative collection time for AED treatment: 24 months after final surgery; postoperative collection time for seizure frequency: from latest follow-up consultation, median (IQR) follow-up of 4.0 years (2.7–5.7) for single-operated patients and 3.3 years (2.6–4.8) for multi-operated patients.
      Abbreviations: IQR: interquartile range; IQ: intelligence quotient; AED: anti-epileptic drug.
      Predictive variables for IQ change in single-operated patients is given in Table 4. In the model adjusted for preoperative IQ only, the preoperative seizure frequency and age at surgery were found to be significant predictors of IQ change. After multivariable adjustment, only preoperative seizure frequency remained a significant predictor of IQ change. In the multivariable analysis, patients with daily seizures preoperatively experienced a smaller overall IQ change postoperatively (-9.9 IQ point change) compared with patients with less than daily seizures, p < 0.01.
      Table 4Predictive variables for IQ change in single-operated patients.
      Individual variable linear regression*Mutually adjusted multivariable linear regression**#
      β-coefficient (95 % CI)

      IQ point change
      p-valueβ-coefficient (95 % CI)

      IQ point change
      p-value
      Female vs. male−5.6 (-11.9 to -0.7)0.08−3.1 (-9.6 to 3.3)0.34
      Left vs. right laterality−3.1 (-9.5 to 3.3)0.33−3.9 (-10.4 to 2.6)0.23
      Age at epilepsy onset (per year)0.8 (-0.1–1.7)0.070.3 (-0.7 to 1.3)0.53
      Age at surgery (per year)0.9 (0.2–1.6)<0.05
      Time from onset of epilepsy to surgery (per year)0.6 (-0.3 to 1.4)0.18
      Preoperative AEDs
      • -
        0
      refref
      • -
        1
      6.4 (-23.3–36.1)0.679.2 (-24.3–42.7)0.58
      • -
        ≥2
      7.1 (-21.9–36.1)0.639.3 (-24.0–42.5)0.58
      Postoperative AEDs
      • -
        0
      refref
      • -
        1
      2.2 (-7.5–11.8)0.66−0.1 (-10.2–10.0)0.99
      • -
        ≥2
      −4.9 (-14.7 to 5.0)0.33−4.1 (-14.7 to 6.5)0.45
      Daily preoperative seizures vs. less than daily preoperative seizures−11.8 (-17.5 to -6.1)<0.001−9.9 (-16.5 to -3.2)<0.01
      Postoperative seizure frequency
      • -
        Seizure-free
      refref
      • -
        Daily
      0.8 (-12.3–13.9)0.905.1 (-10.0–20.3)0.50
      • -
        < daily
      −4.0 (-11.4 to 3.5)0.29−0.6 (-8.5–7.3)0.88
      Postoperative collection time for AED treatment: 24 months after final surgery; postoperative collection time for seizure frequency: from latest follow-up consultation, median (IQR) follow-up of 4.0 years (2.7–5.7) for single-operated patients and 3.3 years (2.6–4.8) for multi-operated patients.
      Abbreviations: AED: anti-epileptic drug.
      *adjusted for preoperative IQ.
      **adjusted for preoperative IQ, gender, laterality, age at epilepsy onset, number of preoperative AEDs, number of postoperative AEDs, preoperative seizure frequency, postoperative seizure frequency.
      #R [
      • Bjørnaes H.
      • Stabell K.
      • Henriksen O.
      • Løyning Y.
      The effects of refractory epilepsy on intellectual functioning in children and adults. A longitudinal study.
      ] = 0.26; F(11, 63) = 2.0, p-value: <0.05.

      3.3 Cognitive function of multi-operated patients

      All except one multi-operated patient underwent at least two postoperative cognitive tests. A total of 10 patients (83.3 %) underwent one-year postoperative cognitive testing, with six (50.0 %) completing one-year postoperative testing more than once. Seven patients (58.3 %) underwent two-year postoperative cognitive testing, with one patient completing two-year postoperative testing more than once. Four patients (33.3 %) underwent both one- and two-year postoperative cognitive testing at various times during their multiple surgeries. Postoperative IQ was calculated from the most recent postoperative IQ test and included five one-year IQ tests and seven two-year IQ tests.
      No significant difference between pre- and postoperative IQ was found among the 12 multi-operated patients, p = 0.46, (Table 2). The number of children with an average intelligence in the normal range preoperatively was 33.3 %, (IQ over 84, Table 1). Three patients had a postoperative increase of ≥10 IQ points, one had a decrease of ≥10 IQ points and eight showed less than a 10-point IQ change at latest postoperative cognitive assessment.

      3.4 Seizure frequency and AED treatment of single-operated patients

      Preoperative seizure frequency and seizure frequency taken from the most recent postoperative assessment is given in Table 2. The median (IQR) follow-up time for seizure frequency was 4.0 years (2.7–5.7). At latest postoperative assessment, 73 patients (92.4 %) showed a decrease in seizure frequency, three patients (3.8 %) experienced an increase in seizure frequency and three patients (3.8 %) experienced no change in seizure frequency. A total of 54 patients (68.4 %) became seizure-free. Overall, a significant number of patients became seizure-free after surgery, p < 0.001.
      At two-year postoperative follow-up, 24 patients (30.4 %) remained on the same number of AEDs as before surgery, 28 patients (35.4 %) were treated with one less AED and 17 patients (21.5 %) with two or more less AEDs, (Table 2). Six patients (7.6 %) received increased AED treatment with the addition of one AED. There was missing data for four patients. Overall, a significant difference between pre- and postoperative AED treatment was found, p < 0.001.

      3.5 Seizure frequency and AED treatment of multi-operated patients

      Preoperative seizure frequency and seizure frequency taken from the most recent postoperative assessment can be seen in Table 2. The median (IQR) follow-up time for seizure frequency was 3.3 years (2.6–4.8). At latest postoperative follow-up, 10 patients (83.3 %) showed a decrease in seizure frequency and two patients (16.7 %) experienced no change in seizure frequency. None experienced an increase in seizure frequency. Eight patients (66.7 %) became seizure-free. Overall, a significant number of patients became seizure-free after surgery, p < 0.01.
      At two-year postoperative follow-up, six patients (50.0 %) remained on the same number of AEDs as before primary surgery, two patients (16.7 %) received one less AED and two patients (16.7 %) received two or more less AEDs. Two patients (16.7 %) received increased AED treatment with the addition of one AED. No significant difference between pre- and postoperative AED treatment was found, p = 0.56, (Table 2).

      4. Discussion

      In this large nationwide cohort study, we examined the pre- and postoperative cognitive function of 91 of 135 consecutive patients during a 20-year period. All included patients underwent focal surgery, with 79 single-operated patients and 12 multi-operated patients.
      IQ remained stable in the majority of single-and multi-operated patients. On a group-level, IQ in single-operated patients did, however, increase significantly and suggests that surgery allows for postoperative IQ gains. Patients operated effectively with a single operation had the best cognitive outcome, where 30 % showed a ≥10-point increase in IQ within one to two years after surgery. Patients who needed more than one operation to achieve a reduction in seizure frequency and/or AED treatment did not show any change in IQ one to two years after the last surgery. These patients did not deteriorate in IQ, as could be expected if surgery had not been performed and suggests that surgery may have a stabilizing effect on cognitive function one to two years postoperatively, even in multi-operated patients. It is important to note, however, that these conclusions are based on findings from a small group of 12 patients. We found a significant postoperative reduction in AED treatment and seizure frequency in single-operated patients. Multi-operated patients did not experience a reduction in AED treatment. Surgery and continued AED treatment did, however, result in two-thirds of these patients becoming seizure-free.
      Our findings of decreased seizure frequency and reduced AED treatment postoperatively are in accordance with previous studies [
      • Skirrow C.
      • Cross J.H.
      • Cormack F.
      • Harkness W.
      • Vargha-Khadem F.
      • Baldeweg T.
      Long-term intellectual outcome after temporal lobe surgery in childhood.
      ,
      • Helmstaedter C.
      • Beeres K.
      • Elger C.E.
      • Kuczaty S.
      • Schramm J.
      • Hoppe C.
      Cognitive outcome of pediatric epilepsy surgery across ages and different types of surgeries: A monocentric 1-year follow-up study in 306 patients of school age.
      ,
      • Wyllie E.
      Surgery for catastrophic localization-related epilepsy in infants.
      ,
      • Dwivedi R.
      • Ramanujam B.
      • Chandra P.S.
      • Sapra S.
      • Gulati S.
      • Kalaivani M.
      Surgery for drug-resistant epilepsy in children.
      ,
      • Mittal S.
      • Montes J.L.
      • Farmer J.-P.
      • Rosenblatt B.
      • Dubeau F.
      • Andermann F.
      Long-term outcome after surgical treatment of temporal lobe epilepsy in children.
      ,
      • Kumar R.M.
      • Koh S.
      • Knupp K.
      • Handler M.H.
      • O’Neill B.R.
      Surgery for infants with catastrophic epilepsy: an analysis of complications and efficacy.
      ,
      • Jenny B.
      • Smoll N.
      • El Hassani Y.
      • Momjian S.
      • Pollo C.
      • Korff C.M.
      Pediatric epilepsy surgery: could age be a predictor of outcomes?.
      ,
      • Meyer F.B.
      • Marsh W.R.
      • Laws E.R.
      • Sharbrough F.W.
      Temporal lobectomy in children with epilepsy.
      ,
      • Puka K.
      • Tavares T.P.
      • Smith M.L.
      Development of intelligence 4 to 11 years after paediatric epilepsy surgery.
      ,
      • Van Oijen M.
      • De Waal H.
      • Van Rijen P.C.
      • Jennekens-Schinkel A.
      • van Huffelen A.C.
      • Van Nieuwenhuizen O.
      Resective epilepsy surgery in childhood: the Dutch experience 1992-2002.
      ,
      • Zupanc M.L.
      • Rubio EJ dos S.
      • Werner R.R.
      • Schwabe M.J.
      • Mueller W.M.
      • Lew S.M.
      Epilepsy surgery outcomes: quality of life and seizure control.
      ,
      • Teutonico F.
      • Mai R.
      • Veggiotti P.
      • Francione S.
      • Tassi L.
      • Borrelli P.
      Epilepsy surgery in children: evaluation of seizure outcome and predictive elements.
      ]. Decreased seizure frequency and reduced need for AED treatment are most likely the result of removal of the epileptogenic focus, but possible surgery-related damage to the blood-brain barrier (BBB), resulting in increased penetrance of AEDs to the brain, has been suggested as a beneficial mechanism [
      • Marchi N.
      • Betto G.
      • Fazio V.
      • Fan Q.
      • Ghosh C.
      • Machado A.
      Blood–brain barrier damage and brain penetration of antiepileptic drugs: role of serum proteins and brain edema.
      ]. On the other hand, a disruption of the BBB may also lead to seizures and treatment-resistant epilepsy [
      • Oby E.
      • Janigro D.
      The blood-brain barrier and epilepsy.
      ,
      • Vezzani A.
      • Friedman A.
      • Dingledine R.J.
      The role of inflammation in epileptogenesis.
      ].
      Regarding cognitive outcome, a review conducted by Van Schooneveld et al. in 2013 pooled data from 16 included studies and found that postoperative IQ/DQ increased in 19 % of the 466 included children, while 11 % experienced a decrease in IQ/DQ and 70 % remained unchanged in IQ/DQ [
      • Van Schooneveld M.M.J.
      • Braun K.P.J.
      Cognitive outcome after epilepsy surgery in children.
      ]. A clear distinction between single- and multi-operated patients was not made and apparently some of the 16 studies included hemispherectomy patients in their cohorts. Our study made the distinction between number of surgeries and we did not include hemispherectomy patients. These differences could have contributed to the low number of patients found with increased IQ/DQ (19 %) when compared with the overall positive IQ increase of our single-operated patients. Since the review by Van Schooneveld, Veersema et al. conducted a study of IQ change in single-operated pediatric patients with mild malformation of cortical development and focal cortical dysplasia. The study showed a ≥10-point IQ increase in 24 % of the 36 included patients one-two years after surgery [
      • Veersema T.J.
      • van Schooneveld M.M.J.
      • Ferrier C.H.
      • van Eijsden P.
      • Gosselaar P.H.
      • van Rijen P.C.
      • et al.
      Cognitive functioning after epilepsy surgery in children with mild malformation of cortical development and focal cortical dysplasia.
      ]. This study closely reflects our findings, where 30 % showed a ≥10-point increase in IQ within one to two years after surgery.
      Due to ethical reasons, it has until recently not been possible to carry out a randomized control study with an appropriate comparison group. Other non-randomized studies have examined the effect of surgery in the form of follow-up studies or by using non-surgery candidates as a comparison group. However, Sibilia et al. recently constructed a case-control study with an appropriate control group. Here, cognitive outcome two years after surgery was examined in 31 DRE pediatric patients who were compared with 14 otherwise identical surgical candidates still awaiting surgery two years after inclusion in the surgical program [
      • Sibilia V.
      • Barba C.
      • Metitieri T.
      • Michelini G.
      • Giordano F.
      • Genitori L.
      Cognitive outcome after epilepsy surgery in children: a controlled longitudinal study.
      ]. An improved developmental trajectory and improved IQ/DQ scores were identifiable only in the surgical group. The study reports very similar overall findings to those we report for our single-operated patients and emphasizes that surgery is an essential component for additional cognitive recovery besides possible IQ improvements due to mere aging of patients.
      Dwivedi et al. constructed a similar randomized control study with a surgery group (operated within a month of randomization) and a medical-therapy group (on a waiting list for surgery), but had different findings from Sibilia et al. [
      • Dwivedi R.
      • Ramanujam B.
      • Chandra P.S.
      • Sapra S.
      • Gulati S.
      • Kalaivani M.
      Surgery for drug-resistant epilepsy in children.
      ]. 30 surgery patients and 33 medical-therapy patients were assessed for IQ change 12 months after inclusion. The surgery group did not experience a change in IQ while the medical-therapy group lost a mean of 3.8 IQ points. No significant difference between the two groups in IQ change was found, but was border-line significant (p = 0.06). Follow-up time in this study may have been too short to identify any possible increase in IQ in the surgery group and the difference between the two groups may have reached significance with longer follow-up, especially if IQ continued to decrease in those awaiting surgery.
      In the multivariable analysis, increased preoperative seizure frequency was the only factor found to be a significant predictor of negative IQ change following surgery. Patients with daily seizures preoperatively had a lower postoperative IQ compared with patients with less than daily preoperative seizures. The fact that preoperative seizure frequency was found to be significant in a multivariable analysis adjusted for a large number of possible confounding factors is compelling evidence of its importance in IQ change postoperatively.
      We found that older age was associated with better postoperative IQ. This is most likely related to later debut of epilepsy. Age at surgery was a significant predictive factor of IQ change in a univariable analysis, but its significance disappeared when examined in a multivariable analysis. If examined in a larger cohort, however, age at surgery may well pertain its significance as an important predictor of IQ change in a multivariable analysis. The same can be said for gender, which neared significance in the univariable analysis, but lost significance in the multivariable analysis. Interestingly, female gender was a negative predictor of positive IQ change. This too, needs to be examined in a larger cohort.
      There are numerous other possible predictors of postoperative cognitive outcome that were not examined in this study and include etiology, histological diagnosis, localization of the lesion, type and location of surgery, completeness of the resection, degree of seizure reduction, postoperative complications, family history of epilepsy, handedness, psychological handicaps and co-morbidities. These possible predictors were not examined either due to issues with multicollinearity, low sample size or were not inventoried in this study. One must conclude that numerous potentially interacting factors can affect cognitive functioning as well as the potential for cognitive improvements in the years following surgery [
      • Berg A.T.
      • Langfitt J.T.
      • Testa F.M.
      • Levy S.R.
      • DiMario F.
      • Westerveld M.
      Global cognitive function in children with epilepsy: a community-based study.
      ,
      • Van Schooneveld M.M.J.
      • Braun K.P.J.
      Cognitive outcome after epilepsy surgery in children.
      ,
      • Ko A.
      • Kim S.H.
      • Kim S.H.
      • Park E.K.
      • Shim K.-W.
      • Kang H.-C.
      • et al.
      Epilepsy surgery for children with low-grade epilepsy-associated tumors: factors associated with seizure recurrence and cognitive function.
      ].
      There are several possible reasons for why multi-operated patients did not show improvement in IQ one to two years after their final surgery. The potential damage inflicted on the brain from repeated surgeries as well as prolonged uncontrolled seizures and treatment with multiple AEDs between surgeries may all require longer recovery times compared with patients operated only once. It is very possible that, with longer follow-up times as well as in a bigger cohort, it may be possible to see improvement in postoperative IQ in this group as well. Also, the multi-operated patients seemed to be a more afflicted group, where age at onset of epilepsy was earlier (median of 2.4 years in multi-operated patients compared with 3.9 years in single-operated patients) and twice as many had daily preoperative seizures compared with single-operated patients (83.3 % of multi-operated patients had daily preoperative seizures versus 46.8 % of single-operated patients).
      The time from inclusion in the epilepsy surgery program to time of surgery was more than one year for both single- and multi-operated patients. Waiting time to surgery did not, however, have a significant effect on IQ outcome in single-operated patients (Table 4). Analyses for multi-operated patients could not be conducted because of low patient numbers. Despite the findings from this study, there is a general consensus that the waiting time to surgery should not be excessively long [
      • Kellermann T.S.
      • Wagner J.L.
      • Smith G.
      • Karia S.
      • Eskandari R.
      Surgical management of pediatric epilepsy: decision-making and outcomes.
      ]. Generally, the shorter the time the brain is exposed to deleterious seizures, the better the postoperative seizure outcome [
      • Hennessy M.J.
      • Elwes R.D.
      • Honavar M.
      • Rabe-Hesketh S.
      • Binnie C.D.
      • Polkey C.E.
      Predictors of outcome and pathological considerations in the surgical treatment of intractable epilepsy associated with temporal lobe lesions.
      ] and the greater the chances for cognitive recovery [
      • Stafstrom C.E.
      • Lynch M.
      • Sutula T.P.
      Consequences of epilepsy in the developing brain: implications for surgical management.
      ,
      • Kellermann T.S.
      • Wagner J.L.
      • Smith G.
      • Karia S.
      • Eskandari R.
      Surgical management of pediatric epilepsy: decision-making and outcomes.
      ].

      4.1 Strengths and limitations

      As in other epilepsy surgery studies, the large heterogeneity of our pediatric epilepsy surgery cohort gave difficulties in comparing patient groups with regard to outcome. This study aimed to be a broad nationwide complete study of all children with epilepsy treated with surgery. Hence, it had broad inclusion criteria and included patients at varying developmental stages with different disease pathologies, variable levels of functioning prior to resection and wide age ranges. Large multicenter studies, possibly from multiple countries, using standardized surgical admission criteria and standardized cognitive tests may be a solution. A strength of this study, however, is the inclusion and assessment of patients from the whole of Denmark going through only two close-working centers, resulting in a surgery cohort with very homogenous (and thus comparable) surgical and cognitive assessments. Furthermore, this study – despite its heterogenic patient group – is one of the largest cohort studies on cognitive function and epilepsy surgery. Also, our cohort and the evolution of our surgical program over the inclusion period is in alignment with neighboring and other industrial countries and, thus, can be applied to a greater setting [
      • Pinborg L.H.
      • Jespersen B.
      • Beniczky S.
      • Fabricius M.
      • Rasonyi G.
      • Uldall P.
      Epilepsy surgery.
      ,
      • Belohlavkova A.
      • Jezdik P.
      • Jahodova A.
      • Kudr M.
      • Benova B.
      • Maulisova A.
      Evolution of pediatric epilepsy surgery program over 2000-2017: improvement of care?.
      ].
      We decided to use a single psychometric measure (IQ/DQ) for cognitive function to counter the heterogeneity and varying performance levels of our cohort. Another contributing factor to our decision was related to the inherent limitations connected to the retrospective nature of this study. Today’s Wechsler Intelligence Tests provide a full-scale IQ as well as verbal comprehension, perceptual reasoning, working memory and processing speed indices. These indices were only available as verbal and performance scores in previous years’ tests. In tests from the beginning of our 20-year inclusion period, only full-scale IQ was available. The decision to only use full-scale IQ prevented us from evaluating possible important differences in specific cognitive domains known to be important for everyday learning and cognitive functioning. Ultimately, we were unable to detect possible fundamental changes in index scores that, when added together, resulted in unchanged full-scale IQ scores.
      A possible learning effect from repeated cognitive testing was not addressed in this study. This is particularly relevant for multi-operated patients, who undergo testing between surgeries and can potentially complete several IQ tests by the time they reach their final surgery. Learning bias should, however, generally not be a problem in children with below average intelligence or severe mental retardation, as was the case for 50 % of our multi-operated patients. Also, since postoperative cognitive testing was moved from one-year follow-up to two-year follow-up in 2013, there should now be a sufficient time gap between tests to counteract a possible learning effect.
      Longer postoperative follow-up times are needed in order to assess the extent of the beneficial effects of surgery on cognitive ability. With the maximum two-year cognitive follow-up time of this study, it was not possible to examine the extent of cognitive recovery and catch-up development and whether there is a limited end level in single-operated patients. It was also not possible to uncover whether multi-operated patients exhibit the same postsurgical pattern of recovery as single-operated patients after an initial stable postoperative phase. A reasonable next-step suggestion would be a five-year postoperative cognitive follow-up.
      Varying data collection times of postoperative outcomes made it challenging to compare patients. Postoperative IQ tests were initially carried out at 1-year follow-up but was later changed to 2-year follow-up. Also, Postoperative AED treatment was noted for all patients at the 2-year postoperative consultation while seizure frequency was noted from the latest follow-up consultation, which for many patients, was outside the follow-up framework of the epilepsy program and several years after final surgery.
      AEDs are known to affect cognitive functioning and adverse cognitive side effects vary according to both drug group and specific medications [
      • Hamed S.A.
      The aspects and mechanisms of cognitive alterations in epilepsy: the role of antiepileptic medications.
      ]. In this study, we did not consider specific AED groups for two primary reasons; this study spans over a 20-year period, where drug treatment options have changed substantially with the introduction of new treatment options over time [

      E. Serrano, A.M. Kanner. Recent treatment advances and novel therapeutic approaches in epilepsy. F1000Prime Rep [Internet]. 26. maj 2015 [henvist 15. maj 2020];7. Tilgængelig hos: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4447056/.

      ]. Furthermore, this study did not have sufficient power to determine any relationship between specific AEDs/drug groups and IQ adequately. The effect of specific AED treatments will need to be examined in a larger, perhaps multi-national study with more recently collected data.
      Finally, despite our homogenous approach to pre- and postoperative evaluations within two close-working centers, patients were assessed by a various people over the evaluation period. This lack of continuity is, of course, a limitation to the study.

      5. Conclusion

      In this nationwide study, we found that resective epilepsy surgery was associated with a beneficial effect on cognitive functioning in pediatric DRE patients. IQ in single-operated patients increased significantly and suggests that surgery allows for postoperative IQ gains. IQ remained stable in multi-operated patients. Increased preoperative seizure frequency was a significant predictor of negative IQ change. Age at surgery was not a significant predictor of IQ change in our multivariable analysis but may well be a significant predictor in larger cohorts. The findings from this study advocate for pediatric epilepsy surgery and we recommend that resective epilepsy surgery be prioritized as a possible treatment option for pediatric patients with focal, structural and drug-resistant epilepsy.

      5.1 Perspectives

      Future perspectives include examining a larger cohort for other predictors of outcome, such as co-morbidities, over longer follow-up times as well as including a non-surgical control group, perhaps consisting of patients on a waiting list for surgery. Also, the separate index scores provided by Wechsler Intelligence Tests as well as other and more specific parameters of cognitive functioning (including various types of memory, psychosocial functioning, academic achievement and behavior) remain to be examined in relation to surgery.

      Funding details

      The study received no external funding.

      Declaration of Competing Interest

      None.

      Author contributions

      PU, CHH, MLB and VK conceived and designed the analysis; VK, KMT, PU, AJ, EM and MK collected the data; VK did the data analysis; VK, CHH and MLB interpreted the results of the data; VK drafted the work; all authors revised the work critically for intellectual content; all authors approved the final version of the work to be published; all authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

      Acknowledgements

      The authors wish to thank the epilepsy surgery group at University Hospital Rigshospitalet and at The Danish Epilepsy Center, Filadelfia and all participating patients and families.

      References

        • Kwan P.
        • Brodie M.J.
        Early identification of refractory epilepsy.
        N Engl J Med. 2000; 342: 314-319
        • Bjørnaes H.
        • Stabell K.
        • Henriksen O.
        • Løyning Y.
        The effects of refractory epilepsy on intellectual functioning in children and adults. A longitudinal study.
        Seizure. 2001; 10: 250-259
        • Tromp S.C.
        • Weber J.W.
        • Aldenkamp A.P.
        • Arends J.
        • vander Linden I.
        • Diepman L.
        Relative influence of epileptic seizures and of epilepsy syndrome on cognitive function.
        J Child Neurol. 2003; 18: 407-412
        • Matsuzaka T.
        • Baba H.
        • Matsuo A.
        • Tsuru A.
        • Moriuchi H.
        • Tanaka S.
        Developmental assessment-based surgical intervention for intractable epilepsies in infants and young children.
        Epilepsia. 2001; 42: 9-12
        • Berg A.T.
        • Langfitt J.T.
        • Testa F.M.
        • Levy S.R.
        • DiMario F.
        • Westerveld M.
        Global cognitive function in children with epilepsy: a community-based study.
        Epilepsia. 2008; 49: 608-614
        • Gordon N.
        Cognitive functions and epileptic activity.
        Seizure. 2000; 9: 184-188
        • Bailet L.L.
        • Turk W.R.
        The impact of childhood epilepsy on neurocognitive and behavioral performance: a prospective longitudinal study.
        Epilepsia. 2000; 41: 426-431
        • Thompson P.J.
        • Duncan J.S.
        Cognitive decline in severe intractable epilepsy.
        Epilepsia. 2005; 46: 1780-1787
        • Kwan P.
        • Arzimanoglou A.
        • Berg A.T.
        • Brodie M.J.
        • Allen Hauser W.
        • Mathern G.
        Definition of drug resistant epilepsy: consensus proposal by the ad hoc Task Force of the ILAE Commission on Therapeutic Strategies.
        Epilepsia. 2010; 51: 1069-1077
        • Bourgeois M.
        • Di Rocco F.
        • Roujeau T.
        • Boddaert N.
        • Lelouch-Tubiana A.
        • Varlet P.
        [Epilepsy and focal lesions in children. Su(1)rgical management].
        Neurochirurgie. 2008; 54: 362-365
        • Freitag H.
        • Tuxhorn I.
        Cognitive function in preschool children after epilepsy surgery: rationale for early intervention.
        Epilepsia. 2005; 46: 561-567
        • Stafstrom C.E.
        • Lynch M.
        • Sutula T.P.
        Consequences of epilepsy in the developing brain: implications for surgical management.
        Semin Pediatr Neurol. 2000; 7: 147-157
        • Holthausen H.
        • Pieper T.
        • Kudernatsch M.
        Towards early diagnosis and treatment to save children from catastrophic epilepsy -- focus on epilepsy surgery.
        Brain Dev. 2013; 35: 730-741
        • Ramantani G.
        • Kadish N.E.
        • Strobl K.
        • Brandt A.
        • Stathi A.
        • Mayer H.
        Seizure and cognitive outcomes of epilepsy surgery in infancy and early childhood.
        Eur J Paediatr Neurol EJPN Off J Eur Paediatr Neurol Soc. 2013; 17: 498-506
        • Elger C.E.
        • Helmstaedter C.
        • Kurthen M.
        Chronic epilepsy and cognition.
        Lancet Neurol. 2004; 3: 663-672
        • Andersson-Roswall L.
        • Engman E.
        • Samuelsson H.
        • Sjöberg-Larsson C.
        • Malmgren K.
        Verbal memory decline and adverse effects on cognition in adult patients with pharmacoresistant partial epilepsy: a longitudinal controlled study of 36 patients.
        Epilepsy Behav EB. 2004; 5: 677-686
        • Cormack F.
        • Cross J.H.
        • Isaacs E.
        • Harkness W.
        • Wright I.
        • Vargha-Khadem F.
        The development of intellectual abilities in pediatric temporal lobe epilepsy.
        Epilepsia. 2007; 48: 201-204
        • Gallagher A.
        • Jambaqué I.
        • Lassonde M.
        Cognitive outcome of surgery.
        Handb Clin Neurol. 2013; 111: 797-802
        • Van Schooneveld M.M.J.
        • Braun K.P.J.
        Cognitive outcome after epilepsy surgery in children.
        Brain Dev. 2013; 35: 721-729
        • Gleissner U.
        • Sassen R.
        • Schramm J.
        • Elger C.E.
        • Helmstaedter C.
        Greater functional recovery after temporal lobe epilepsy surgery in children.
        Brain J Neurol. 2005; 128: 2822-2829
        • Skirrow C.
        • Cross J.H.
        • Cormack F.
        • Harkness W.
        • Vargha-Khadem F.
        • Baldeweg T.
        Long-term intellectual outcome after temporal lobe surgery in childhood.
        Neurology. 2011; 76: 1330-1337
        • Wyllie E.
        • Comair Y.G.
        • Kotagal P.
        • Raja S.
        • Ruggieri P.
        Epilepsy surgery in infants.
        Epilepsia. 1996; 37: 625-637
        • Boshuisen K.
        • van Schooneveld M.M.J.
        • Uiterwaal C.S.P.M.
        • Cross J.H.
        • Harrison S.
        • Polster T.
        Intelligence quotient improves after antiepileptic drug withdrawal following pediatric epilepsy surgery.
        Ann Neurol. 2015; 78: 104-114
        • Liu S.
        • An N.
        • Yang H.
        • Yang M.
        • Hou Z.
        • Liu L.
        Pediatric intractable epilepsy syndromes: reason for early surgical intervention.
        Brain Dev. 2007; 29: 69-78
        • Loddenkemper T.
        • Holland K.D.
        • Stanford L.D.
        • Kotagal P.
        • Bingaman W.
        • Wyllie E.
        Developmental outcome after epilepsy surgery in infancy.
        Pediatrics. 2007; 119: 930-935
        • Helmstaedter C.
        • Beeres K.
        • Elger C.E.
        • Kuczaty S.
        • Schramm J.
        • Hoppe C.
        Cognitive outcome of pediatric epilepsy surgery across ages and different types of surgeries: A monocentric 1-year follow-up study in 306 patients of school age.
        Seizure. 2019; 26
        • Westerveld M.
        • Sass K.J.
        • Chelune G.J.
        • Hermann B.P.
        • Barr W.B.
        • Loring D.W.
        Temporal lobectomy in children: cognitive outcome.
        J Neurosurg. 2000; 92: 24-30
        • Roulet-Perez E.
        • Davidoff V.
        • Mayor-Dubois C.
        • Maeder-Ingvar M.
        • Seeck M.
        • Ruffieux C.
        Impact of severe epilepsy on development: recovery potential after successful early epilepsy surgery.
        Epilepsia. 2010; 51: 1266-1276
        • Wyllie E.
        Surgery for catastrophic localization-related epilepsy in infants.
        Epilepsia. 1996; 37: S22-25
        • Pinborg L.H.
        • Jespersen B.
        • Beniczky S.
        • Fabricius M.
        • Rasonyi G.
        • Uldall P.
        Epilepsy surgery.
        Ugeskr Laeger. 2018; 180
        • von Celsing Underbjerg E.
        • Hoei-Hansen C.E.
        • Madsen F.F.
        • Madsen C.G.
        • Høgenhaven H.
        • Uldall P.
        Danish experience with paediatric epilepsy surgery.
        Dan Med J. 2015; 62: A5164
        • Dwivedi R.
        • Ramanujam B.
        • Chandra P.S.
        • Sapra S.
        • Gulati S.
        • Kalaivani M.
        Surgery for drug-resistant epilepsy in children.
        N Engl J Med. 2017; 26: 1639-1647
        • Mittal S.
        • Montes J.L.
        • Farmer J.-P.
        • Rosenblatt B.
        • Dubeau F.
        • Andermann F.
        Long-term outcome after surgical treatment of temporal lobe epilepsy in children.
        J Neurosurg. 2005; 103: 401-412
        • Kumar R.M.
        • Koh S.
        • Knupp K.
        • Handler M.H.
        • O’Neill B.R.
        Surgery for infants with catastrophic epilepsy: an analysis of complications and efficacy.
        Childs Nerv Syst ChNS Off J Int Soc Pediatr Neurosurg. 2015; 31: 1479-1491
        • Jenny B.
        • Smoll N.
        • El Hassani Y.
        • Momjian S.
        • Pollo C.
        • Korff C.M.
        Pediatric epilepsy surgery: could age be a predictor of outcomes?.
        J Neurosurg Pediatr. 2016; 18: 235-241
        • Meyer F.B.
        • Marsh W.R.
        • Laws E.R.
        • Sharbrough F.W.
        Temporal lobectomy in children with epilepsy.
        J Neurosurg. 1986; 64: 371-376
        • Puka K.
        • Tavares T.P.
        • Smith M.L.
        Development of intelligence 4 to 11 years after paediatric epilepsy surgery.
        J Neuropsychol. 2015; 17
        • Van Oijen M.
        • De Waal H.
        • Van Rijen P.C.
        • Jennekens-Schinkel A.
        • van Huffelen A.C.
        • Van Nieuwenhuizen O.
        Resective epilepsy surgery in childhood: the Dutch experience 1992-2002.
        Eur J Paediatr Neurol EJPN Off J Eur Paediatr Neurol Soc. 2006; 10: 114-123
        • Zupanc M.L.
        • Rubio EJ dos S.
        • Werner R.R.
        • Schwabe M.J.
        • Mueller W.M.
        • Lew S.M.
        Epilepsy surgery outcomes: quality of life and seizure control.
        Pediatr Neurol. 2010; 42: 12-20
        • Teutonico F.
        • Mai R.
        • Veggiotti P.
        • Francione S.
        • Tassi L.
        • Borrelli P.
        Epilepsy surgery in children: evaluation of seizure outcome and predictive elements.
        Epilepsia. 2013; 54: 70-76
        • Marchi N.
        • Betto G.
        • Fazio V.
        • Fan Q.
        • Ghosh C.
        • Machado A.
        Blood–brain barrier damage and brain penetration of antiepileptic drugs: role of serum proteins and brain edema.
        Epilepsia. 2009; 50: 664-677
        • Oby E.
        • Janigro D.
        The blood-brain barrier and epilepsy.
        Epilepsia. 2006; 47: 1761-1774
        • Vezzani A.
        • Friedman A.
        • Dingledine R.J.
        The role of inflammation in epileptogenesis.
        Neuropharmacology. 2013; 69: 16-24
        • Veersema T.J.
        • van Schooneveld M.M.J.
        • Ferrier C.H.
        • van Eijsden P.
        • Gosselaar P.H.
        • van Rijen P.C.
        • et al.
        Cognitive functioning after epilepsy surgery in children with mild malformation of cortical development and focal cortical dysplasia.
        Epilepsy Behav. 2019; 94: 209-215
        • Sibilia V.
        • Barba C.
        • Metitieri T.
        • Michelini G.
        • Giordano F.
        • Genitori L.
        Cognitive outcome after epilepsy surgery in children: a controlled longitudinal study.
        Epilepsy Behav EB. 2017; 73: 23-30
        • Ko A.
        • Kim S.H.
        • Kim S.H.
        • Park E.K.
        • Shim K.-W.
        • Kang H.-C.
        • et al.
        Epilepsy surgery for children with low-grade epilepsy-associated tumors: factors associated with seizure recurrence and cognitive function.
        Pediatr Neurol. 2019; 91: 50-56
        • Kellermann T.S.
        • Wagner J.L.
        • Smith G.
        • Karia S.
        • Eskandari R.
        Surgical management of pediatric epilepsy: decision-making and outcomes.
        Pediatr Neurol. 2016; 64: 21-31
        • Hennessy M.J.
        • Elwes R.D.
        • Honavar M.
        • Rabe-Hesketh S.
        • Binnie C.D.
        • Polkey C.E.
        Predictors of outcome and pathological considerations in the surgical treatment of intractable epilepsy associated with temporal lobe lesions.
        J Neurol Neurosurg Psychiatry. 2001; 70: 450-458
        • Belohlavkova A.
        • Jezdik P.
        • Jahodova A.
        • Kudr M.
        • Benova B.
        • Maulisova A.
        Evolution of pediatric epilepsy surgery program over 2000-2017: improvement of care?.
        Eur J Paediatr Neurol EJPN Off J Eur Paediatr Neurol Soc. 2019; 23: 456-465
        • Hamed S.A.
        The aspects and mechanisms of cognitive alterations in epilepsy: the role of antiepileptic medications.
        CNS Neurosci Ther. 2009; 15: 134-156
      1. E. Serrano, A.M. Kanner. Recent treatment advances and novel therapeutic approaches in epilepsy. F1000Prime Rep [Internet]. 26. maj 2015 [henvist 15. maj 2020];7. Tilgængelig hos: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4447056/.